Neuroblastoma, one of the most common solid tumors in chilren, has been treated with limited success with various regimens of cytotoxic chemotherapeutic drugs. It is known to spontaneously differentiate in vivo perhaps preventing tumors from presenting themselves in many children. In model studies, mouse neuroblastoma cells grown in tissue culture have been treated with dibutyryl cAMP and other agents increasing intracellular cAMP levels. These cells are not only biochemically and morphologically differentiated but also are less tumorigenic in the syngeneic Ajax mouse. We propose to investigate the correlation between differentiation in clonal lines of these cells, as evidenced by increases in the levels of the specific neuronal markers, choline acetyltransferase and tyrosine hydroxylase, and their in vivo tumorigenicity. The transcriptional, translational and post-translational mechanisms underlying this differentiation will be studied at several levels. We shall study the effectors of transcriptional control, the nuclear acidic proteins and the changes in the overall synthesis and processing of nuclear RNAs. Control of cytoplasmic expression at the translational level will be studied by the analysis of changes in levels of the in vivo synthesis of specific proteins, separated by two dimensional gel electrophoresis. These in vivo labeling studies will be extended by analysis of the levels and subcellular localization of cytoplasmic messenger RNAs, assayed by the proteins they synthesize in an in vitro protein synthesizing system. We will also investigate the synthesis and/or glycosylation of membrane proteins as these may be the direct modulators of a cell's tumorigenicity. Having identified these cellular alterations which positively correlate with reduced tumorigenicity, we will then investigate their mechanism, defining a system in which to predict and test a new class of non-toxic drugs capable of differentiating neuroblastoma in vivo.